With the development of Internet technology and the diversification of service, various types of content services have been provided over the Internet. In particular, services using multimedia content, such as videos or advertisements, beyond the service of viewing various types of images over the Internet, have considerably increased. Furthermore, recently, with the rapid popularization of smart phones, cases where images and videos are provided over the Internet and a mobile communication network via smart phones have increased. Meanwhile, the provision and utilization of images as well as videos over a network have also increased. For example, images acquired by capturing the topical scenes of a popular soap opera or show program being broadcast are instantaneously distributed via the Internet notice-boards of a number of news sites, blogs, cafes, etc. immediately after the broadcast of the program. Furthermore, advertising images or images directly captured by users are being provided via SNS services, communities, the notice-boards of portal sites, etc. The reason for this is that posts including images can attract the attention of users more than text-centric posts.
As described above, it is not easy to associate images with videos or provide various types of associated additional services based on images in spite of a situation in which an enormous number of images are being provided over the Internet. The reason for this is that there is no method capable of identifying videos or other additional services corresponding to images over a network in real time.
For example, a user may view a news article, including an image acquired by capturing a scene of a soap opera, over the Internet, and then may want to immediately watch the scene including the corresponding image via a video. In this case, currently, the user should additionally search a web site that provides the corresponding video. Furthermore, even in such a case, the scene desired by the user is not accurately provided, and thus the user should personally search the provided video for the scene desired by him or her.
In this regard, there is a need for a method capable of providing videos or various types of additional services to client terminals based on images distributed over a network, such as the Internet or a mobile communication network, in an automated manner or in real time.
Meanwhile, Korean Patent Application Publication No. 10-2009-0096158 (published on Sep. 10, 2009) discloses a “Video Still Cut-based Cross-Media Search System and Method.” In this publication, there are disclosed a method and system for receiving a still cut, that is, a search target, extracting image features by analyzing the input still cut, searching for associated webpage information based on the extracted image feature information by referring to a webpage database (DB) containing the per-frame image feature information of video still cuts included in meta information about webpages and webpages themselves, and then providing found webpage information.
However, this preceding technology is merely a technology for simply searching for and providing a webpage or video information corresponding to an input image, and has its limitations in terms of accurate searching for a video and a frame corresponding to an image based on the image provided by a client terminal and the provision of a video service corresponding to the corresponding video and frame.